INDEX

Explanations

phrases and references related to research studies and reports

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

istine

-0.15

or

-0.14

 bars

-0.14

sul

-0.14

ragon

-0.14

et

-0.14

rub

-0.14

otti

-0.14

 earlier

-0.13

POSITIVE LOGITS

ë§¥

0.15

lsen

0.15

¹

0.14

aber

0.14

-FIRST

0.14

piring

0.14

 interv

0.13

 tÃŃnh

0.13

Ð°Ð

0.13

oga

0.13

Activations Density 0.064%

phrases and references related to research studies and reports

No Comments

No Known Activations